1. Field of the Invention
This invention relates to the field of vehicle identification. More particularly, this invention relates to use of sound in identifying and tracking vehicles, and methods based on identification of the vehicles using acoustic machine signatures.
2. Background
Mechanical systems, such as vehicles, frequently are identified for a variety of reasons. Commonly, vehicle identification and tracking relies upon visually-related data, such as make/model information or license plate data. At times, such as in low-light or no-light situations, gathering visually-related data may be difficult or impossible. Also, visually-related data such as license plates or vehicle color may be changed to deflect efforts to identify a vehicle. Once visually-related data has been changed, visual vehicle identification may be difficult or impossible.
Typically, unless significant mechanical change or wear occurs, an “acoustic signature” or “machine signature” generated during operation of the vehicle generally remains consistent over the life of the vehicle. The acoustic signature includes the sound of the engine in operation as well as the sound of the wheels of the vehicle traversing a known surface. Discussions of identification of vehicles using acoustic signatures may be found in: B. Guo et al., “Acoustic Information Fusion for Ground Vehicle Classification”, 11th International Conference on Information Fusion, pp. 1-7, Jun. 30, 2008-Jul. 3, 2008, Cologne, Germany (“Guo”), M. E. Munich, “Bayesian Subspace Methods for Acoustic Signature Recognition of Vehicles”, Proceedings of the 12th European Signal Processing Conference (EUSIPCO), pp. 2107-2110, Sep. 6-10, 2004, Vienna, Austria (“Munich”), and B. Lu et al., “Perimeter Security on Detecting Acoustic Signature of Approaching Vehicle Using Nonlinear Neural Computation”, 2008 IEEE Conference on Technologies for Homeland Security, pp. 51-56, May 12-13, 2008, Waltham, Mass. (“Lu”). Guo, Munich, and Lu are all entirely incorporated by reference herein for all purposes.